How does Bolt predict mileage ranges?

scbolt · Jul 4, 2017

After a full charge recently, my Bolt was showing this:

Was curious how the Bolt decides the mileage values like max/nominal/min (374/317/259 above). Does it predict this based on recent battery usage and efficiency data?

I have seen the max value to be as high as even 400 miles and was wondering how that is possible.

grestep5621 · Jul 4, 2017

I think it seems to be a simple calculation of the miles per kWh and the kWh left in the battery.
It doesn't seem to vary until my miles per kWh changes.
I drive fairly consistently, but as I have become more familiar with the Bolt I drive more efficiently and the miles per kWh is going up.
I drove the first 1000 miles without using L and now rarely use D anymore.
L driving is much more efficient and I basically know now when to let up on the accelerator to achieve a stop at most distances.
Other than the miles per kWh I don't see any other factor that determines what you see on the SOC gauge until you start driving.

scbolt · Jul 4, 2017

grestep5621 said:
I think it seems to be a simple calculation of the miles per kWh and the kWh left in the battery.
It doesn't seem to vary until my miles per kWh changes.
I drive fairly consistently, but as I have become more familiar with the Bolt I drive more efficiently and the miles per kWh is going up.
I drove the first 1000 miles without using L and now rarely use D anymore.
L driving is much more efficient and I basically know now when to let up on the accelerator to achieve a stop at most distances.
Other than the miles per kWh I don't see any other factor that determines what you see on the SOC gauge until you start driving.

Yes, very similar experience here. I wasn't driving in L mode until about a month back ... and now I drive 100% in L mode. My mi/kWh was floating around 3.8 but after a recent trip up and down Mt. Evans in CO (~14,300 ft) I got amazing regen (on the way down) which pushed up my average, and subsequently the Bolt's internal calculation of available mileage.

I am sure it is a simple calculation of mi/kWh and capacity left in the battery ... but my main question is how does it predict the mi/kWh? Does it simply use the mi/kWh from the last trip? Or in the current trip so far? Or some other average. It certainly does not use lifetime average, as you can see from my screenshot (my lifetime avg is 4.0 mi/kWh so far) -- since in my case that will give around 240 miles or so (60 x 4.0). So it must use some more recent data.

gbobman · Jul 5, 2017

It's using "some other average" that they have not explained yet. Probably some sort of rolling average over a week or 2. It has already been dubbed the GOM or Guess O' Meter.

scbolt · Jul 5, 2017

gbobman said:
It's using "some other average" that they have not explained yet. Probably some sort of rolling average over a week or 2. It has already been dubbed the GOM or Guess O' Meter.

Aha! So that is what "GOM" is

... thanks for explaining! Saw this mentioned several times here but was not sure.

sgt1372 · Jul 5, 2017

My estimated range after every recharge just seems to be equal to the average m/khw shown on the screen x 60 kwh (the battery capacity of the Bolt). Just charged the car today. Average m/kwh was 3.1 and the estimated range after charging was 186. Exactly 3.1 x 60.

BTW, is it just me or does anyone else notice that after recharging their Bolt that the average m/kwh always starts out at 3.9, which equals 234 or just about the 238 mile advertised range for the car, regardless of what the actual average use has been previously?

EldRick · Jul 6, 2017

BTW, is it just me or does anyone else notice that after recharging their Bolt that the average m/kwh always starts out at 3.9, which equals 234 or just about the 238 mile advertised range for the car, regardless of what the actual average use has been previously?

It's just you and your reliable weather. I've seen as low as 151 on the GoM after a full charge.

gbobman · Jul 6, 2017

sgt1372 said:
My estimated range after every recharge just seems to be equal to the average m/khw shown on the screen x 60 kwh (the battery capacity of the Bolt). Just charged the car today. Average m/kwh was 3.1 and the estimated range after charging was 186. Exactly 3.1 x 60.

BTW, is it just me or does anyone else notice that after recharging their Bolt that the average m/kwh always starts out at 3.9, which equals 234 or just about the 238 mile advertised range for the car, regardless of what the actual average use has been previously?

When I hit reset on the trip interface, yes, always 3.9 immediately. Then goes right up to 4.1 when I start off where I normally sit with these two cheap tire replacements I have...

sgt1372 · Jul 7, 2017

gbobman said:
When I hit reset on the trip interface, yes, always 3.9 immediately. Then goes right up to 4.1 when I start off where I normally sit with these two cheap tire replacements I have...

My average also readjusts to however I'm driving after the recharge. Currently, it's a 3.4 because all I did was in-town driving after the most recent recharge but any trip on the freeway will quickly drop it back to 3.1 or so.

scbolt · Jul 7, 2017

EldRick said:
BTW, is it just me or does anyone else notice that after recharging their Bolt that the average m/kwh always starts out at 3.9, which equals 234 or just about the 238 mile advertised range for the car, regardless of what the actual average use has been previously?

Click to expand...

It's just you and your reliable weather. I've seen as low as 151 on the GoM after a full charge.

I used to see very consistent reading on the GoM after a full charge, but it DRASTICALLY increased after last Saturday, when we had gone to the mountains and got massive regen on the way back, which increased my average mi/kWh stat. Hence I was trying to figure out how exactly the GoM calculates the predicted range ... and how long of a "history" it uses

gbobman · Jul 7, 2017

scbolt said:
I used to see very consistent reading on the GoM after a full charge, but it DRASTICALLY increased after last Saturday, when we had gone to the mountains and got massive regen on the way back, which increased my average mi/kWh stat. Hence I was trying to figure out how exactly the GoM calculates the predicted range ... and how long of a "history" it uses

It still may be a rolling average, just with a huge spike in it to skew the averages. I am hoping we do find out though, since spikes like that can really throw of planning if we aren't paying much attention.

joe · Jul 18, 2017

I believe I can shed some light on this question.

TL;DR: I think it uses an average from the history of the last 100 miles.

I did a long tour this weekend on high-speed (70+) highways with AC, and my average efficiency went way down (mid to low 4s in mi/kWh). I did a full charge and the estimated range was about 280. I resumed my normal commute immediately after the full charge, and after resetting the trip odometer/efficiency calculator, it took about 15-20 miles to get back up to the usual 6ish mi/kWh from where it started at 3.9. (I have also noticed it starts at 3.9 after I reset the trip odometer). I've driven about 140 miles since the full charge and resetting the trip odometer, and for the first 100 miles, the estimated range was stuck around 270 +/- 10mi. It was not until about almost exactly 100 miles into the "trip" that the range meter started to steadily decrease below 260. Since that time, the estimated range and the trip odometer distance have been adding up to the expected range given the trip efficiency, which at 6 mi/kWh is 360, whereas before, the est. range+trip distance just kept climbing.

It makes sense to me that they would pick some magic number like "100 miles" for efficiency average to calculate the estimated range. It should be large enough so there aren't wild fluctuations in the estimate, but small enough that it reflects recent trends.

scbolt · Jul 18, 2017

joe said:
I believe I can shed some light on this question.

TL;DR: I think it uses an average from the history of the last 100 miles.

I did a long tour this weekend on high-speed (70+) highways with AC, and my average efficiency went way down (mid to low 4s in mi/kWh). I did a full charge and the estimated range was about 280. I resumed my normal commute immediately after the full charge, and after resetting the trip odometer/efficiency calculator, it took about 15-20 miles to get back up to the usual 6ish mi/kWh from where it started at 3.9. (I have also noticed it starts at 3.9 after I reset the trip odometer). I've driven about 140 miles since the full charge and resetting the trip odometer, and for the first 100 miles, the estimated range was stuck around 270 +/- 10mi. It was not until about almost exactly 100 miles into the "trip" that the range meter started to steadily decrease below 260. Since that time, the estimated range and the trip odometer distance have been adding up to the expected range given the trip efficiency, which at 6 mi/kWh is 360, whereas before, the est. range+trip distance just kept climbing.

It makes sense to me that they would pick some magic number like "100 miles" for efficiency average to calculate the estimated range. It should be large enough so there aren't wild fluctuations in the estimate, but small enough that it reflects recent trends.

this is great info! thanks.

Anonymous · Jul 22, 2017

gbobman said:
It's using "some other average" that they have not explained yet. Probably some sort of rolling average over a week or 2. It has already been dubbed the GOM or Guess O' Meter.

The "Guess o Meter" moniker was given to the Nissan Leaf and is an insult to the Bolt EV!

like2bike · Jul 22, 2017

I don't understand why Nissan, GM, etc can't fill us in on what the estimate is based on. It's a difficult thing to reverse engineer. I'm glad to hear the Bolt's estimate is based on a much longer period than the LEAF's GOM. I quickly found the GOM annoying as the range would plummet as soon as I started going up a hill. 100 miles seems like a much better range period. What is really needed is a nav system which predicts the range based on all factors and real-time info. available. When are we going to have an e-Waze which serves all the needs of EVs? Presently, I'll be happy when Waze acknowledges EVs have different needs and starts providing for them.

Anonymous · Jul 22, 2017

The most accurate way is to watch the Energy Used as it increases to 60 and watch the mileage remaining on your trip and know the remaining topology.

Thats how I drive to Northern California and back over the Grapevine

SeanNelson · Jul 22, 2017

MichaelLAX said:
The "Guess o Meter" moniker was given to the Nissan Leaf and is an insult to the Bolt EV!

While I appreciate the sentiment given the wonky way the Leaf's estimated range works, even the Bolt's estimated range is still just a guess - so "Guess-O-Meter" is an appropriate moniker, IMHO.

Anonymous · Jul 22, 2017

SeanNelson said:
MichaelLAX said:

The "Guess o Meter" moniker was given to the Nissan Leaf and is an insult to the Bolt EV!

Click to expand...

While I appreciate the sentiment given the wonky way the Leaf's estimated range works, even the Bolt's estimated range is still just a guess - so "Guess-O-Meter" is an appropriate moniker, IMHO.

Apparently the algorithm used in the Leaf was not much more than a true "guess!"

On the contrary, in the Bolt EV, the Mileage Gauge is an accurate measure of your past performance.

In other words, if you drove the exact same trip as you did the day before under the same conditions, the Mileage Gauge would be an exact predictor of the mileage remaining as your completed the trip.

There is no guessing involved and using the word "guess" to describe the Mileage Gauge on the Bolt EV is an insult to all Bolt EV drivers!

But hey, your mileage may differ!!! hahahaha

WetEV · Jul 23, 2017

MichaelLAX said:
There is no guessing involved and using the word "guess" to describe the Mileage Gauge on the Bolt EV is an insult to all Bolt EV drivers!

"Guess" is a caution that the mileage gauge is a projection based on history, and does not, as far as anyone knows, take into account today's conditions, other than temperature.

Traffic varies. Where you drive varies. How you drive varies. Wind varies. Snow, rain and so forth.

It might be a better guess, but don't forget that it is just a guess.

It is a projection, nothing more.

GOM is a fairly old term, and predates modern electric cars. The range estimate is a GOM.

SeanNelson · Jul 23, 2017

MichaelLAX said:
There is no guessing involved and using the word "guess" to describe the Mileage Gauge on the Bolt EV is an insult to all Bolt EV drivers!

There certainly is a guess involved, because the car cannot predict future conditions. Even if you do drive exactly the same route, traffic, weather conditions and other factors can and do affect range. It is very unwise to place too much trust in the estimated range, particularly if you know that your future driving conditions are going to differ from past ones (you need to go up a long grade, for example).

Yes, it's a much more educated guess than we've seen in some other cars. But an educated guess it still is.

Using "Guess-O-Meter" to colourfully describe the estimated range reminds us of the need to be cautious about it, and it's also a good way to help educate new EV drivers who have a consistent habit of asking why they can't get the advertised range or even the range that's displayed in big numbers on their dashboards.

How does Bolt predict mileage ranges?

Help Support Chevy Bolt EV Forum:

Well-known member

Active member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Anonymous

Guest

Well-known member

Anonymous

Guest

Well-known member

Anonymous

Guest

Well-known member

Well-known member

Similar threads

Join the conversation!

Join today and get all the highlights of this community direct to your inbox. It's FREE!